Economic dispatch (ED) is one of the most important optimization problems in a power system. The objective of ED is sharing the power demand among the online generators while keeping the minimum cost of generation as a constraint. The aim of this paper is to operate an electric power system as economically as possible within its security limits. This paper proposes the following 2 new particle swarm optimization (PSO) algorithms to solve a nonconvex economic dispatch problem: an efficient PSO is termed as efficient particle swarm optimization (EPSO), and a hybrid of evolutionary programming (EP) and EPSO is termed as EP-EPSO. Since ED was introduced, several methods have been used to solve these problems. However, none of these methods can provide an optimal solution because they become trapped at some local optima. Stochastic optimization techniques such as EPSO and EP have the advantage of a good convergent property. A significant speed-up can be obtained by the hybrid of these algorithms. The proposed techniques are tested on standard test systems available in the literature. The performance of the proposed EP-EPSO is compared with a) biogeography-based optimization, b) adaptive particle swarm optimization, c) the genetic algorithm, d) a 2-phase neural network, e) PSO with time-varying acceleration coefficients, f) NEW-PSO, and g) differential evolution with biogeography-based optimization. It is observed that the EP-EPSO has a higher convergence rate, advanced quality, and better optimal cost when compared to the other techniques. The considered ED problems have been solved, including transmission losses without valve-point loading effects.

Economic dispatch (ED) is one of the most important optimization problems in a power system. The objective of ED is sharing the power demand among the online generators while keeping the minimum cost of generation as a constraint. The aim of this paper is to operate an electric power system as economically as possible within its security limits. This paper proposes the following 2 new particle swarm optimization (PSO) algorithms to solve a nonconvex economic dispatch problem: an efficient PSO is termed as efficient particle swarm optimization (EPSO), and a hybrid of evolutionary programming (EP) and EPSO is termed as EP-EPSO. Since ED was introduced, several methods have been used to solve these problems. However, none of these methods can provide an optimal solution because they become trapped at some local optima. Stochastic optimization techniques such as EPSO and EP have the advantage of a good convergent property. A significant speed-up can be obtained by the hybrid of these algorithms. The proposed techniques are tested on standard test systems available in the literature. The performance of the proposed EP-EPSO is compared with a) biogeography-based optimization, b) adaptive particle swarm optimization, c) the genetic algorithm, d) a 2-phase neural network, e) PSO with time-varying acceleration coefficients, f) NEW-PSO, and g) differential evolution with biogeography-based optimization. It is observed that the EP-EPSO has a higher convergence rate, advanced quality, and better optimal cost when compared to the other techniques. The considered ED problems have been solved, including transmission losses without valve-point loading effects.